Data Assimilation and Analysis PE 0602435 (NRL BE-035-02-19)

LONG-TERM GOALS: To build data assimilation systems that analyze all scales of weather, within both central site and on-scene computer environments. This project will deliver a highly effective data assimilation system that will produce the best linear unbiased estimate of the atmosphere's state at a particular time, using a prediction model, observations from a 24 hour period, and the most recent analysis. The ultimate system will take a number of years, yet deliverables will be made on components of this system so that the operational data assimilation system will continue to improve throughout the research period. Near-term deliverables include the newest version of the mesoscale data assimilation system, the one-dimensional variational method to assimilate satellite radiances directly, and the capability to assimilate rain-rates into the global prediction system. The end result will be more precise data assimilation systems, based on sound optimal estimation theory, which will lead to better, longer-term weather forecasts by the various prediction systems. The impact will be greatest in the ocean areas due to improved utilization of satellite and on-scene sensors. OBJECTIVES: To develop a fully modern three-dimensional variational method capable of assimilating measured quantities of non-conventional and remotely sensed observations in addition to the conventional values of temperature, wind, and moisture that are currently processed. To improve methods of computing error characteristics of the various components of the data assimilation suite and the observations, and to develop methods to detect errors in both the data and the prediction system. To design and build systems and support software that use modern local area networks, database technology, and graphics libraries, to better display and diagnose important features of the environment. To exploit research and development at other weather prediction centers and laboratories worldwide.